A comprehensive review highlights the variability in defining developmental regression across studies, calling for standardised criteria to improve diagnosis, research, and support for affected children.
Researchers have found that developmental regression is being studied far more often than before, but the field still lacks a consistent way to define or measure it. A scoping review in the journal Autism Research examined 157 peer-reviewed studies published between 1963 and April 2025 and concluded that while most papers used some form of operational definition, the details varied widely, especially around when regression began, how long it lasted and which skills were lost.
That inconsistency matters because developmental regression is not a single condition but a feature seen across autism, developmental epileptic encephalopathies, genetic syndromes and neurodegenerative disorders. The review found that autism dominated the literature, accounting for more than three-quarters of the studies, while smaller numbers focused on progressive neurodegenerative diseases, epilepsy-related conditions and genetic disorders such as Rett syndrome and Phelan-McDermid syndrome. Across these groups, the presentation of regression differed enough to make a universal definition difficult, but also enough to make one increasingly necessary.
The review said the literature has expanded sharply since 2000, with most papers appearing after 2010. It also showed that researchers most commonly relied on parent reports and medical records rather than a shared gold-standard tool. The Autism Diagnostic Interview-Revised was the most frequently used structured measure, but the authors noted that it was often paired with other instruments and that only a small minority of studies used regression-specific tools designed to capture the loss of skills more directly.
The authors argued that this patchwork approach has practical consequences. Different definitions can change who is included in a study, distort prevalence estimates and make it harder to compare findings across groups. They can also delay diagnosis and support, especially when a child’s regression is recorded only as part of a broader diagnosis rather than as a distinct clinical problem. The review said this is particularly important because children with autism who experience regression often have greater support needs than those without it.
The paper also highlighted a wider trend in clinical research: as awareness of regression has grown, so has the need for better phenotype definitions and more careful measurement. The authors said future work should pay closer attention to pre-loss abilities, the timing of regression, the duration of skill loss and the use of home video as an objective source of evidence. They concluded that a standard definition would strengthen research quality and improve recognition, treatment and family support.
Disclaimer: This content is for informational purposes only and is not intended to be a substitute for professional medical judgment, advice, diagnosis, or treatment.
